A stratified sequencing multi-disciplinary reliability analysis method under random and interval uncertainty

For the reliability analysis problems in multidisciplinary systems with stochastic and interval uncertainty models in coexistence, based on the sequence of variable processing framework, the original problem is decomposed into the interval reliability analysis and probabilistic reliability analysis sub-problems, and the corresponding solution procedure is established. Among them, interdisciplinary consistency constraints are introduced into the interval reliability analysis to reduce the calculation burden of multidisciplinary coupling analysis; advanced mean value method, arc search method and active set quasi Newton method are integrated into reliability analysis process to construct a set of asymptotic convergence processing strategy. Finally, we build the iterative relationship between the two sub-processes based on the idea of tolerance hierarchical structure and process random and interval uncertainty simultaneously. A numerical example and a reliability analysis application in multidisciplinary system of flight vehicle are demonstrated to verify the validity of the proposed method from different aspects. Simulation results show that the efficiency is increased and adaptability is rather strong in the mode of various initial values and design points.